TY - JOUR
AU - Shai, Meiri,
AB - Abstract Animal lifespan is determined by extrinsic and intrinsic factors causing mortality. According to the evolutionary theories of senescence, when mortality pressures are low, animals delay reproduction. This enables species to grow more slowly and, consequently, natural selection can act against harmful mutations in adulthood, thereby increasing lifespans. To test predictions of these theories we assembled a dataset on the maximum longevities and relevant ecological variables of 1320 reptilian species. Correcting for phylogeny, we modelled the link between reptile longevity and factors such as body size, microhabitat, activity period, insularity, annual temperature, temperature seasonality, elevation and clutch size that we hypothesized will affect extrinsic mortality rates and hence lifespan. Body mass explained a small proportion of the variance in reptile longevity. Species living on islands, and in colder and more seasonal environments, lived longer. Observed maximum longevity was positively associated with the number of individuals used to estimate it. Our results suggest that species exposed to reduced extrinsic and intrinsic mortality pressures (lower predation, lower metabolic rates and shorter activity periods) live longer. Sampling more individuals increases the chances of finding older specimens and should be corrected for when studying maximum longevity. body size, comparative analysis, evolutionary theories of senescence, longevity, phylogeny, seasonal temperature INTRODUCTION Most organisms die as a consequence of exposure to external factors such as diseases, parasitism, social interactions, famine, accidents or predation (Wachter & Finch, 1997). Those that escape such direct factors ultimately face death through intrinsic processes of senescence, such as spontaneous chemical reactions, replication errors and accumulation of metabolic waste products (Koopman et al., 2015). Consequently, mortality has been attributed to either intrinsic or extrinsic causes (Sparkman et al., 2007; Shattuck & Williams, 2010). Senescence, a decline in fitness with age, can be explained by a decrease in the force of natural selection with age (Medawar, 1952; Williams, 1957; Kirkwood & Rose, 1991). Other theories consider ageing to be a physiological process that evolved largely independently of ecological factors, pointing to an association between metabolism and progressive intrinsic deterioration of the body (Sohal & Allen, 1986; Finch, 1990). Those evolutionary theories of senescence predict that members of populations exposed to high levels of extrinsic mortality will evolve short potential lifespans (Healy et al., 2014; Valcu et al., 2014). Because most individuals die from extrinsic causes, natural selection cannot act on reducing intrinsic causes of mortality, which otherwise lead to longer lifespans (Williams, 1957; Hamilton, 1966; Williams & Day, 2003; Williams et al., 2006; Caswell, 2007). High extrinsic mortality rates select for early reproduction, placing organisms under additional stress, which can, in turn, increase intrinsic mortality (Quinlan, 2010). Interspecific and intraspecific comparisons have been used to test whether adaptations that presumably reduce the risk of intrinsic and extrinsic mortality are associated with longer lifespans (Valcu et al., 2014). Most studies that focus on the relationship between longevity and the factors related to mortality have been conducted on birds and mammals (e.g. Lindstedt & Calder, 1981; Holmes & Austad, 1994; Wilkinson & South, 2002; Speakman, 2005; Turbill et al., 2011; Healy et al., 2014; Valcu et al., 2014). Stark & Meiri (2018) found that factors such as the use of venom and nocturnality are associated with longer lifespans in amphibians. They ascribed this to reduced predation pressure (low extrinsic mortality rate). Scharf et al. (2015) analysed the lifespan records of 1014 squamate species, and consistent with the predictions of the evolutionary theory of senescence, their results suggest that factors such as herbivory, living in regions with low environmental productivity and living at higher latitudes result in longer lifespans, which they ascribed mostly to reduction in intrinsic mortality rates. Scharf et al. (2015), however, mostly concentrated on factors co-varying along a fast–slow life-history continuum (i.e. living on the slower side of the continuum results in longer lifespan, because of delayed reproduction), rather than examining the impact of mortality factors. Furthermore, when alluding to mortality factors they mostly examined the effects of intrinsic and not of extrinsic mortality factors (on which we focus here). Moreover, Scharf et al. (2015) did not examine either testudines (here signifying all chelonians) or crocodilians, while we include both orders, and use a larger and more comprehensive dataset for squamates as well. Here we study the effects of factors thought to be related to extrinsic mortality rates on the longevity of reptiles from all orders. In accordance with the evolutionary theories of senescence (Medawar, 1952; Williams, 1957; Kirkwood & Rose, 1991), regarding the effects of environmental and life-history components on longevity, we concentrate on several key traits. Microhabitat: we predict longevity to be highest for fossorial species, then for arboreal species and then for species of all other microhabitat uses (e.g. terrestrial, saxicolous, aquatic). This is because predation pressure (and hence lower extrinsic mortality) will be lower for arboreal, aquatic, especially, fossorial reptiles (Buffenstein & Jarvis, 2002; Shattuck &Williams, 2010; Byrnes & Spence, 2011; Healy et al., 2014). Activity period: nocturnal reptiles are less exposed to solar radiation, and experience lower temperatures than diurnal species, consequently reducing their metabolic rates. This, in turn, leads to lower intrinsic mortality rates (Sohal & Allen, 1986). Ambient temperature: the metabolic rate of ectotherms is often low at low temperature (regions at high latitudes and altitudes: Sohal, 1986; Zhang & Lu, 2012). Thus, intrinsic mortality agents are suspected to operate more slowly in such regions. Moreover, species living in colder, more seasonal areas experience short activity seasons, and are less exposed to predators during hibernation, reducing extrinsic mortality pressure and perhaps leading to longer lifespans (Turbill et al., 2011; Meiri et al., 2013). Insularity: species living on islands often face few predator, competitor and/or pathogen species. They may thus be less exposed to extrinsic mortality pressures (Austad, 1993; Jordana et al., 2012). Defence mechanisms: venomous snakes and testudines (because of their shell) may experience lower predation pressure than other reptile species (Martín et al., 2005; Fry et al., 2006; Jansa & Voss, 2011; Mori et al., 2012; Hossie et al., 2013). Moreover, venomous snakes usually handle their prey less than non-venomous species (Downes, 2002), lowering the probability of injuries. We predict that short-lived species will have larger clutch size, on average. Species suffering higher mortality pressures may invest more in reproduction, producing more offspring per reproductive event. METHODS Data collection We collected literature data on the maximum longevity of 1320 reptile species spanning all reptilian orders (and 67 of 87 families): Squamata (745 lizards including five amphisbaenid species and 353 snake species), Testudines (192 turtle and tortoise species), Crocodilia (all 24 crocodile species) and Sphenodontia (the tuatara, Sphenodon punctatus). This represents c. 12.3% of known reptile species diversity (currently 10793 species: Uetz et al., 2018). Data on the predictor variables were gathered from the same sources used to estimate longevity and, if not reported there, from the literature (see Supporting Information, Appendix S1). Longevity data are the maximum lifespan (in years) reported for each species. Unfortunately, there are some drawbacks in using maximum values as estimates of longevity. For example, maxima may be unrepresentative of the species as a whole (Carey, 2003), or estimates could be based on unequal numbers of individuals for different species, whereas means are probably unbiased (Scharf et al., 2015). Therefore, we recorded the sample sizes upon which longevity is estimated for each species to correct for these potential biases inherent in the use of maxima (i.e. maxima are expected to increase with sample size; Meiri, 2007). We further introduced a sensitivity analysis for the 531 species for which adequate sample sizes (N ≥ 30) were reported (for other species, samples were smaller or data on sample sizes were not reported), and repeated our analysis of the other predictors. We further analysed crocodiles, testudines and squamates separately, to determine whether this basic split of the reptile tree of life influences the traits affecting longevity. We used data on maximum body masses of reptiles from Slavenko et al. (2016). Masses are based on maximum length data that were converted to mass via allometric equations for different reptile clades (mostly family-level equations, taking leg development data into account when this varied within a family; Feldman & Meiri, 2013; Feldman et al., 2016). Because we focus on maximum longevity, and reptiles usually grow throughout their lives, it is reasonable to expect that the oldest individuals will also be among the largest. Maxima are likewise the most common of size measures available for reptiles (e.g. Meiri, 2008; Slavenko et al., 2016). Mean sizes, in contrast, are more representative measures than maxima. We collected data on mean body size (length converted to mass as above). When multiple means were available from different sources of literature, we calculated the mean of the highest and lowest published means of males and females. We performed a sensitivity analysis using mean mass to test whether the use of maxima for mass biases our results on the relationship between body size and longevity. We use mass because it is a better size index than length, as it accounts for variation in shape (Meiri, 2010; Feldman & Meiri, 2014; Feldman et al., 2016). Mass differs greatly between animals of similar length that are as different from each other as colubrid snakes and tortoises. We classified species as diurnal, nocturnal or cathemeral. We categorized crepuscular species (N < 15) as diurnal, because they are exposed to high daytime temperatures and to solar radiation and thus presumably have high metabolic rates (Buchanan, 1998; Stark & Meiri, 2018). We classified species as terrestrial, arboreal, fossorial (including semi-fossorial) or aquatic. Some species are active in more than one type of habitat (e.g. arboreal and terrestrial or saxicolous and fossorial), and we treat them as ‘variable’. For snakes, we collected data on the possession of venom, and classify snakes as either venomous, rear-fanged or non-venomous (Blanco & Sherman, 2005; Hossie et al., 2013). We classified species into insular endemics or continental based on distribution data from Roll et al. (2017). We compared the longevity of mainland species to that of insular endemics across all islands, and also to species inhabiting only on islands smaller than 50000 km2, then 5000 km2 and finally 1000 km2. Mean annual temperatures and temperature seasonality were estimated for each species by intersecting the latitude and longitude of the location where longevity data (from the wild) were collected with a layer of mean annual temperature and temperature seasonality within 30 arc seconds from Fick & Hijmans (2017) (WorldClim database). Such coordinates, however, were only available for 125 species. For the other species we averaged climatic data (WorldClim) across all the grid-cells they inhabit, according to ranges from Roll et al. (2017). We also collected data for the elevation range of 70% of the species in our dataset. We used the mid-point of the elevational range as another predictor (using either the minimum or the maximum elevation reached by a species yielded qualitatively identical results; data not shown). For clutch or litter size we use species means from Meiri (2018) for lizards, Itescu (2012) for turtles, and Feldman (2015) for snakes. For crocodiles and Sphenodon we obtained the data from the same sources listing longevity (Appendix S1). If means were unavailable we used range midpoints (e.g. the average between the largest and smallest known clutches). Similarly, we use a midpoint of the largest and smallest mean if multiple means were reported. We tested if longevities are affected by whether data were from captive or wild individuals. While captive specimens are not exposed to extrinsic mortality factors such as predation, we reason that their maximum longevity reflects a life-history strategy that evolved in response to selection pressures in the wild. Nonetheless, captive individuals may have higher longevity because they are not preyed upon, receive ample food, etc. To further examine this we directly compared species for which we had longevity data from both captive and wild individuals (N = 352) using a paired t-test. We further introduced a sensitivity analysis in which we only use species for which maximum longevity was recorded in the wild, and re-ran our model for all reptiles on this subset (227 species). In all other analyses we used the highest longevity datum for each species. Phylogenetic analyses We combined the most up-to-date and complete phylogenetic trees for each order. For squamates we used a 9574-species tree from Tonini et al. (2016). For testudines we used a 294-species tree from Pereira et al. (2017). We reconstructed a Bayesian inference multilocus tree of Crocodilia using BEAST v.1.8.4 (Drummond et al., 2012), with sequences from four mitochondrial regions and nine nuclear loci obtained from Oaks (2011). Parameters and priors, as well as the resulting tree, are presented in Appendix S2. We combined the trees, using the methods and R scripts outlined in Roquet et al. (2014). We used the ‘caper’ package in R (Orme et al., 2013) to estimate the maximum likelihood value of Pagel’s λ, a statistically powerful index for measuring whether data exhibit phylogenetic dependence and how strong it is (Freckleton et al., 2002). We accounted for shared ancestry using phylogenetic generalized least square (PGLS) tests, adjusting the strength of phylogenetic non-independence using the maximum likelihood value of the scaling parameter value λ (Pagel, 1999), implemented in the R package ‘caper’ (Orme et al., 2013). Statistical analyses We log10-transformed data on maximum body mass, maximum longevity, clutch size and sample size to normalize residuals and reduce heteroscedasticity. We calculated variance inflation factors (VIFs) to determine whether multicollinearity between the predictor variables was present in the data. No predictor had a VIF higher than 2.5 and we thus maintain that our analyses are not biased by multicollinearity (O’Brien, 2007). We obtained data on the predictor variables for 70% (for elevation) to 100% (body size) of the species in our database. We conducted a phylogenetic ANCOVA test starting from a full model and using a backwards stepwise elimination procedure, sequentially deleting factors with P-values > 0.005 (Johnson, 2013; Benjamin et al., 2018) until arriving at a minimum adequate model where all predictors were significant (at α < 0.005). We compared the allometric PGLS equations (slopes, intercepts) of longevity vs. body size between the three orders (excluding Sphenodontia) using t-tests. Finally, we determined the importance of the selected predictor variables by variance partitioning. All statistical analyses were performed in R 3.3.2 (R Core Team, Vienna, Austria) using RStudio (1.1.383, RStudio Inc., Boston, MA, USA). RESULTS Mean (±SE) longevity across all reptile orders we analysed was 17.1 ± 0.5 years, with maximum longevity ranging from 1 year (Menetia greyii, Pianka, 2011) to 186 years (Aldabrachelys gigantea, living in captivity on St. Helena Island, Castanet, 1994). Modal reptile longevity was 10 years (Fig. 1). For squamates species average maximum longevity was 12.6 ± 0.3 years (range: 1–69, N = 1103), for testudines 36.5 ± 2.3 years (range 3.3–186, N = 192) and for crocodilians 62.9 ± 4.1 years (range 21.1–100, N = 24); the tuatara can live at least 120 years. Figure 1. View largeDownload slide Distribution of the (log10) maximum longevity (in years) of the 1320 reptilian species. Figure 1. View largeDownload slide Distribution of the (log10) maximum longevity (in years) of the 1320 reptilian species. Reptile longevity increases to the 0.19 power of mass (99.5% confidence interval = 0.161–0.218). The slope was steepest for lizards, but much shallower for the other taxa (Table 1). Crocodile longevity was not correlated with body size (P = 0.140). The interaction between the slopes was significant (P = 0.001), but pairwise differences were only significant for the difference between lizards and testudines (P = 0.005, Fig. 2). Overall, maximum body size explained just 4% (for testudines) to 23% (for lizards) of the variance in reptilian longevity. We obtained similar results when using mean (N = 1297 species) and maximum mass (1320 species) both among and within reptile clades (see Table A1 in Appendix S3.1). Analysis of species with large sample sizes showed similar values (Table A2 in Appendix S3.1). Table 1. Longevity as a function of body mass for reptiles (phylogenetic analyses) Confidence interval of slope P N λ Intercept ± 1 SE t Slope ± 1 SE R2 Group 0.161–0.218 < 0.0001 1305 0.800 0.739 ± 0.182 14.6 0.190 ± 0.013 0.14 Reptilia 0.241–0.298 < 0.0001 746 0.644 0.448 ± 0.094 14.8 0.270 ± 0.012 0.23 Lacertilia 0.074–0.131 < 0.0001 351 0.784 0.852 ± 0.144 5.6 0.103 ± 0.014 0.08 Serpentes 0.006–0.233 0.004 183 0.561 0.918 ± 0.202 2.8 0.120 ± 0.041 0.04 Testudines −0.126 to 0.372 0.140 24 0.000 1.109 ± 0.427 1.5 0.123 ± 0.080 0.09 Crocodilia Confidence interval of slope P N λ Intercept ± 1 SE t Slope ± 1 SE R2 Group 0.161–0.218 < 0.0001 1305 0.800 0.739 ± 0.182 14.6 0.190 ± 0.013 0.14 Reptilia 0.241–0.298 < 0.0001 746 0.644 0.448 ± 0.094 14.8 0.270 ± 0.012 0.23 Lacertilia 0.074–0.131 < 0.0001 351 0.784 0.852 ± 0.144 5.6 0.103 ± 0.014 0.08 Serpentes 0.006–0.233 0.004 183 0.561 0.918 ± 0.202 2.8 0.120 ± 0.041 0.04 Testudines −0.126 to 0.372 0.140 24 0.000 1.109 ± 0.427 1.5 0.123 ± 0.080 0.09 Crocodilia View Large Table 1. Longevity as a function of body mass for reptiles (phylogenetic analyses) Confidence interval of slope P N λ Intercept ± 1 SE t Slope ± 1 SE R2 Group 0.161–0.218 < 0.0001 1305 0.800 0.739 ± 0.182 14.6 0.190 ± 0.013 0.14 Reptilia 0.241–0.298 < 0.0001 746 0.644 0.448 ± 0.094 14.8 0.270 ± 0.012 0.23 Lacertilia 0.074–0.131 < 0.0001 351 0.784 0.852 ± 0.144 5.6 0.103 ± 0.014 0.08 Serpentes 0.006–0.233 0.004 183 0.561 0.918 ± 0.202 2.8 0.120 ± 0.041 0.04 Testudines −0.126 to 0.372 0.140 24 0.000 1.109 ± 0.427 1.5 0.123 ± 0.080 0.09 Crocodilia Confidence interval of slope P N λ Intercept ± 1 SE t Slope ± 1 SE R2 Group 0.161–0.218 < 0.0001 1305 0.800 0.739 ± 0.182 14.6 0.190 ± 0.013 0.14 Reptilia 0.241–0.298 < 0.0001 746 0.644 0.448 ± 0.094 14.8 0.270 ± 0.012 0.23 Lacertilia 0.074–0.131 < 0.0001 351 0.784 0.852 ± 0.144 5.6 0.103 ± 0.014 0.08 Serpentes 0.006–0.233 0.004 183 0.561 0.918 ± 0.202 2.8 0.120 ± 0.041 0.04 Testudines −0.126 to 0.372 0.140 24 0.000 1.109 ± 0.427 1.5 0.123 ± 0.080 0.09 Crocodilia View Large Figure 2. View largeDownload slide The relationship between maximum longevity (log-transformed) and body mass (log-transformed) of the major reptilian groups for the complete dataset (N = 1320): Lacertilia (green circles), Serpentes (black triangles), Testudines (purple diamonds) and Crocodilia (blue squares). Sphenodon punctatus is marked by a red cross. Figure 2. View largeDownload slide The relationship between maximum longevity (log-transformed) and body mass (log-transformed) of the major reptilian groups for the complete dataset (N = 1320): Lacertilia (green circles), Serpentes (black triangles), Testudines (purple diamonds) and Crocodilia (blue squares). Sphenodon punctatus is marked by a red cross. Longevity decreased in warm regions (slope = −0.006 ± 0.002) and increased in seasonal regions (slope = 0.001 ± 0.001). Insular species (N = 226) lived, on average, 12.1% longer than mainland species (N = 1092). Insular species lived longer than mainland ones even when we examined only species endemic to islands smaller than 50000, 5000 and 1000 km2 (see Tables A3–A5 in Appendix 3.2). Values of maximum longevity were positively correlated with the number of specimens on which they were estimated (slope = 0.082 ± 0.007). While these results are in agreement with our hypotheses, other factors (microhabitat, activity period, clutch size, elevation and origin of data) were non-significantly associated with longevity (P > 0.05), contradicting our predictions. The best model explained 22% of the variation in longevity, with body size (partial R2 = 37.2%) explaining the largest part of the overall variance in longevity (λ = 0.79, N = 1061, see Table 2 for effect sizes and other statistics). For the full model see Table A3 in Appendix S3.3. The minimum adequate model for species with longevity data originating only from the wild (N = 227 species) was qualitatively similar to that for all species (i.e. it contained most of the same significant predictors with the same signs; see Appendix 3.4). Consequently, we only refer to models using the full dataset (1320 species) below. Table 2. The minimal adequate model for the full reptile dataset and tree Partial R2 (%) P t SE Estimate Factor NA < 0.0001 5.3 0.182 0.0.982 Intercept 37.2 < 0.0001 9.3 0.013 0.124 Body mass (log grams) 1.3 0.0051 -2.7 0.002 -0.006 Mean annual temperature 1.3 0.0054 2.7 0.001 0.001 Temperature seasonality 2.1 0.001 3.2 0.026 0.083 Insularity 5.4 < 0.0001 10.7 0.007 0.082 Sample size (log10) Partial R2 (%) P t SE Estimate Factor NA < 0.0001 5.3 0.182 0.0.982 Intercept 37.2 < 0.0001 9.3 0.013 0.124 Body mass (log grams) 1.3 0.0051 -2.7 0.002 -0.006 Mean annual temperature 1.3 0.0054 2.7 0.001 0.001 Temperature seasonality 2.1 0.001 3.2 0.026 0.083 Insularity 5.4 < 0.0001 10.7 0.007 0.082 Sample size (log10) Estimates for body mass, mean annual temperature, temperature seasonality and sample size are slopes, while the estimate for insularity is an intercept. The first row is the intercept for mainland species. For species living on islands the intercept is calculated by adding the estimate value in the corresponding row to this value (0.982). t and P values for insularity refer to differences from species living on the mainland. R2 = 0.22, N = 1061, λ = 0.798, P < 0.0001. View Large Table 2. The minimal adequate model for the full reptile dataset and tree Partial R2 (%) P t SE Estimate Factor NA < 0.0001 5.3 0.182 0.0.982 Intercept 37.2 < 0.0001 9.3 0.013 0.124 Body mass (log grams) 1.3 0.0051 -2.7 0.002 -0.006 Mean annual temperature 1.3 0.0054 2.7 0.001 0.001 Temperature seasonality 2.1 0.001 3.2 0.026 0.083 Insularity 5.4 < 0.0001 10.7 0.007 0.082 Sample size (log10) Partial R2 (%) P t SE Estimate Factor NA < 0.0001 5.3 0.182 0.0.982 Intercept 37.2 < 0.0001 9.3 0.013 0.124 Body mass (log grams) 1.3 0.0051 -2.7 0.002 -0.006 Mean annual temperature 1.3 0.0054 2.7 0.001 0.001 Temperature seasonality 2.1 0.001 3.2 0.026 0.083 Insularity 5.4 < 0.0001 10.7 0.007 0.082 Sample size (log10) Estimates for body mass, mean annual temperature, temperature seasonality and sample size are slopes, while the estimate for insularity is an intercept. The first row is the intercept for mainland species. For species living on islands the intercept is calculated by adding the estimate value in the corresponding row to this value (0.982). t and P values for insularity refer to differences from species living on the mainland. R2 = 0.22, N = 1061, λ = 0.798, P < 0.0001. View Large Analysing those species with large samples (> 30 specimens per species, 531 species), we obtained somewhat similar results, with two exceptions: temperature seasonality (P = 0.106) and insularity (P = 0.093) were not associated with longevity. Body mass explained the largest part (~45%) of the overall variance in longevity in this dataset (see Appendix S3.5). The three reptile orders were characterized by different models. The models for squamates were very similar to the full reptilian model (i.e. most of the same factors were significant, with similar signs, see Table 3 for effect sizes and other statistics, and Table A4 in Appendix S3.3 for the full model), with one exception: mean annual temperature had a marginally significant effect on longevity (P = 0.03). The large sample size database for squamates (> 30 specimens per species, 424 species) was likewise qualitatively the same as that for all reptiles (see Appendix S3.5). Table 3. The minimal adequate model for Squamata only Partial R2 (%) P t SE Estimate Factor NA < 0.0001 5.0 0.095 0.485 Intercept 31.0 < 0.0001 11.3 0.016 0.186 Body mass (log grams) 6.9 < 0.0001 5.3 0.003 0.002 Temperature seasonality 3.5 0.001 3.1 0.026 0.083 Insularity 9.1 < 0.0001 9.9 0.008 0.081 Sample size (log) Partial R2 (%) P t SE Estimate Factor NA < 0.0001 5.0 0.095 0.485 Intercept 31.0 < 0.0001 11.3 0.016 0.186 Body mass (log grams) 6.9 < 0.0001 5.3 0.003 0.002 Temperature seasonality 3.5 0.001 3.1 0.026 0.083 Insularity 9.1 < 0.0001 9.9 0.008 0.081 Sample size (log) Estimates for body mass, mean annual temperature, temperature seasonality and sample size are slopes, while the estimate for insularity is an intercept. The first row is the intercept for mainland species. For species living on islands the intercept is calculated by adding the estimate value in the corresponding row to this value (0.485). t and P values for insularity refer to differences from species living on the mainland. R2 = 0.25, N = 885, λ = 0.715, P < 0.0001. View Large Table 3. The minimal adequate model for Squamata only Partial R2 (%) P t SE Estimate Factor NA < 0.0001 5.0 0.095 0.485 Intercept 31.0 < 0.0001 11.3 0.016 0.186 Body mass (log grams) 6.9 < 0.0001 5.3 0.003 0.002 Temperature seasonality 3.5 0.001 3.1 0.026 0.083 Insularity 9.1 < 0.0001 9.9 0.008 0.081 Sample size (log) Partial R2 (%) P t SE Estimate Factor NA < 0.0001 5.0 0.095 0.485 Intercept 31.0 < 0.0001 11.3 0.016 0.186 Body mass (log grams) 6.9 < 0.0001 5.3 0.003 0.002 Temperature seasonality 3.5 0.001 3.1 0.026 0.083 Insularity 9.1 < 0.0001 9.9 0.008 0.081 Sample size (log) Estimates for body mass, mean annual temperature, temperature seasonality and sample size are slopes, while the estimate for insularity is an intercept. The first row is the intercept for mainland species. For species living on islands the intercept is calculated by adding the estimate value in the corresponding row to this value (0.485). t and P values for insularity refer to differences from species living on the mainland. R2 = 0.25, N = 885, λ = 0.715, P < 0.0001. View Large In testudines, the minimum adequate model revealed a negative relationship between longevity and mean annual temperature, and a positive relationship between longevity and sample size. In contrast to our predictions, however, we found that terrestrial testudines (N = 43 species) live, on average, 34.5% longer than aquatic testudines (N = 117). This model explained 37% of the variation in longevity, with mean annual temperature (15.8%) explaining the largest part (see Table 4 for effect sizes and other statistics, and Table A5 in Appendix S3.3 for the full model). Insular testudine species live longer, on average, than their mainland counterparts (17.7% longer lifespan; corrected for size). The phylogenetic signal was zero in this particular analysis, and seasonality (P = 0.984) did not affect longevity as it did for Squamata and for reptiles in general. The large sample size database for testudines resulted in a qualitatively similar model (see Appendix S3.5), with two exception: terrestrial testudines live longer than aquatic ones, but this result is not very strong (P = 0.02). Sample size was uncorrelated with longevity (P = 0.159). For crocodilians, the phylogenetic signal was likewise zero and no predictors were significantly associated with longevity (P > 0.05 for all predictors), but note that sample size was much smaller (all 24 extant species). We found the same (negative) results in the large sample-size dataset (N = 10 species) of crocodiles. For a summary of all results for all reptile groups see Appendix S3.6. Table 4. The minimal adequate model for Testudines Partial R2 (%) P t SE Estimate Factor NA < 0.0001 14.5 0.128 1.872 Intercept 8.3 0.0005 −3.5 0.056 −0.199 Microhabitat (aquatic) 12.1 < 0.0001 −4.8 0.005 −0.024 Mean annual temperature 5.1 0.004 2.6 0.094 0.250 Insularity 18.6 < 0.0001 4.9 0.020 0.101 Sample size (log) Partial R2 (%) P t SE Estimate Factor NA < 0.0001 14.5 0.128 1.872 Intercept 8.3 0.0005 −3.5 0.056 −0.199 Microhabitat (aquatic) 12.1 < 0.0001 −4.8 0.005 −0.024 Mean annual temperature 5.1 0.004 2.6 0.094 0.250 Insularity 18.6 < 0.0001 4.9 0.020 0.101 Sample size (log) Estimates for body mass, mean annual temperature and sample size are slopes, while estimates for microhabitat and insularity are intercepts. The first row is the intercept for terrestrial species living on the mainland. For species that feed in the water and living on islands the intercept is calculated by adding the estimate value in the corresponding row to this value (1.872). t and P values for microhabitat refer to differences from species feeding only on the ground. R2 = 0.35, N = 157, λ = 0.000, P < 0.0001. View Large Table 4. The minimal adequate model for Testudines Partial R2 (%) P t SE Estimate Factor NA < 0.0001 14.5 0.128 1.872 Intercept 8.3 0.0005 −3.5 0.056 −0.199 Microhabitat (aquatic) 12.1 < 0.0001 −4.8 0.005 −0.024 Mean annual temperature 5.1 0.004 2.6 0.094 0.250 Insularity 18.6 < 0.0001 4.9 0.020 0.101 Sample size (log) Partial R2 (%) P t SE Estimate Factor NA < 0.0001 14.5 0.128 1.872 Intercept 8.3 0.0005 −3.5 0.056 −0.199 Microhabitat (aquatic) 12.1 < 0.0001 −4.8 0.005 −0.024 Mean annual temperature 5.1 0.004 2.6 0.094 0.250 Insularity 18.6 < 0.0001 4.9 0.020 0.101 Sample size (log) Estimates for body mass, mean annual temperature and sample size are slopes, while estimates for microhabitat and insularity are intercepts. The first row is the intercept for terrestrial species living on the mainland. For species that feed in the water and living on islands the intercept is calculated by adding the estimate value in the corresponding row to this value (1.872). t and P values for microhabitat refer to differences from species feeding only on the ground. R2 = 0.35, N = 157, λ = 0.000, P < 0.0001. View Large Snake longevity was not associated with possessing venom (λ = 0.769, N = 347, P = 0.02 for the contrast between non-venomous and rear-fanged species, and P = 0.78 for non-venomous vs. fully venomous species). Interestingly, when we tested (using PGLS) the differences in longevity between testudines and other reptile groups, we did not find differences. A non-phylogenetic analysis (corrected for sample size) revealed that small testudines species live longer than other reptile species, but the difference diminishes with increasing sizes, and are reversed for reptiles >63 kg (R2 = 0.490, N = 1077, P = 0.003). For effect sizes and other statistics see Appendix 3.7. A paired t-test revealed no intraspecific differences in longevity between captive and wild individuals (N = 352 species; average longevity in the wild: 20.8 ± 1.2 years, in captivity: 20.8 ± 1.1 years, t = 0.01, P = 0.99, see Appendix S3.8). DISCUSSION Our comparative analysis of lifespans of 1320 species across all reptilian orders supports some predictions derived from the evolutionary theories of senescence, but for others, the results are contradicting or equivocal. Reptile longevity is enhanced for large species, those living on islands, and those exposed to lower ambient and highly seasonal temperatures. Other factors, however, such as activity period, venomousness or captivity did not affect longevity either in reptiles in general or within any order. Across most of our models, body size was positively correlated with lifespan (expect for the low-power crocodile models), consistent with previous studies (e.g. Austad & Fischer, 1991; Healy et al., 2014; Scharf et al., 2015; Holm et al., 2016; Stark & Meiri, 2018). Large species have a slower metabolism, which translates to slower growth (Speakman, 2005). This is associated with a reduction in the rate of accumulation of metabolic by-products and oxidative damage (Sohal, 1986). Consequently, large size can result in lower exposure to extrinsic mortality factors. Body mass, however, explained little of the variance in reptile longevity (R2 = 14%). This may be because ectotherms are less affected by metabolic by-products because of their lower metabolic rates (Bronikowski & Vleck, 2010). Additionally, strong effects of temperature on ectotherms may mean that size has less variation to affect, but we did not find such a strong thermal signal in our analyses. For a given increase in body mass, the lifespans of testudines species (slope: 0.11 ± 0.03) increase slowly. Small-bodied testudines thus live much longer than similar-sized squamates, but the differences diminish with increasing size, and are reversed for the largest squamates (> 63 kg). This is probably due to the shell playing a highly beneficial role in reducing predation on small testudines, but it may be less effective, and more costly to carry and maintain, at larger sizes. Thus, a ~600-g Chrysemys dorsalis turtle can live to 20.6 years compared to the similar-sized (575-g) Leiolepis guttata lizard that reaches 13 years (examples chosen from species close to their respective regression lines). Larger testudines, however, seem to have a similar lifespan to other large reptiles (although admittedly such large sizes are rare in squamates). This is despite the fact that some of the largest testudines are herbivorous, insular, land tortoises (Itescu et al., 2014); these are additional factors that could promote longevity. The relationships among the ecological variables we examined and lifespan differed among and within reptilian clades. Temperature was negatively related to reptilian longevity, a pattern that was shown in other ectothermic invertebrate and vertebrate groups such as crustaceans, gastropods, fish and amphibians (Munch & Salinas, 2009; Stark & Meiri, 2018). Furthermore, lifespan is longer in more seasonal regions. This may be the result of a faster growth rate in climatically stable regions, leading to a faster accumulation of harmful metabolic by-products. Such a relationship has been linked to the rate-of-living theory, which posits that lifespan is dependent upon the exhaustion of a fixed quantity of a vital substance at a rate proportional to the metabolic rate (Sohal, 1986). A similar link is predicted in the oxidative damage theory of ageing, which posits that animals that produce more free radicals suffer more molecular damage, which in turn should lead to short lives (Brys et al., 2007). Additionally, in colder, more seasonal environments reptiles are active for shorter periods of the year (Johnston & Dunn, 1987; Meiri et al., 2013), thereby reducing their predation risk and risk of food shortages (Turbill et al., 2011). The longer lifespans of insular reptilian species support a model suggesting slower life-history traits on islands (often called ‘island syndrome’; Adler & Levins, 1994; Grant, 1998; Novosolov et al., 2013). Higher longevity on islands can be the result of lower exposure to fewer predator species and pathogens, which may result in lower extrinsic mortality (Jordana et al., 2012). Sample size was positively correlated with longevity in most of our analyses. It appears that increasing sample size strongly increases the probability of finding older individuals. We urge that sample size be controlled for in future comparative analyses of maximum longevity. It seems the different microhabitats influenced longevity in testudines, but not in other reptile groups. Consequently, we suggest that the different microhabitats do not help reduce predation rates (lower extrinsic mortality). Surprisingly, however, even fossorial species, which we reasoned must enjoy relaxed predation pressures (Williams & Shattuck, 2015), did not have longer lifespans. Thus, either this assumption is incorrect (e.g. if species are frequently preyed upon in the short periods they spend above ground, or even hunted below ground) as shown for mammals (when corrected for phylogeny and sociality; Healy et al., 2014), or the metabolic costs associated with burrowing can cancel out the advantages of lower predation pressures. Activity period was likewise not related to lifespan in reptile species, refuting our initial prediction. It may be that high daytime temperatures in the regions inhabited by nocturnal reptiles (Vidan et al., 2017) may increase their metabolic rates when they are not active, and that solar radiation is of little effect. Alternatively (but less parsimoniously), increased predation rates at night (Daly et al., 1992) may offset lower metabolic rates and solar radiation experienced by nocturnal species. It may even be that cellular repair mechanisms are less efficient at the lower temperatures at which nocturnal reptiles operate. We have no direct data on predation rates, radiation effects or the efficacy of cellular repair mechanisms to allow us to test any of these hypotheses. Interestingly, and in contradiction to the results of both Scharf et al. (2015) for squamates and for Stark & Meiri (2018) for amphibians, captivity did not influence the recorded longevity within species. We cannot easily explain this discrepancy. We tentatively suggest that factors such as stress, poor keeping conditions and disease may cause premature deaths in captivity, offsetting the positive effects of reducing predation and hunger (Robinson et al., 2015). We expected short-lived species to produce more offspring per clutch in order to maintain long-term persistence of the species. However (and in contradiction to the results of Scharf et al., 2015), this expectation was not met by our results. We suggest that by analysing a more diverse assemblage of reptiles together, multiple reproductive strategies mask such a relationship. The use/possession of venom did not help to decrease extrinsic mortality in snakes. Hossie et al. (2013) showed a similar pattern to ours with a smaller sample (N = 102). This may suggest that producing venom is costly (McCue, 2006; Smith et al., 2014), negating beneficial effects of reduced predation. Alternatively, predators may suffer from the same misconceptions that many humans have, and may be unable to distinguish between potentially venomous and non-venomous snakes. Moreover, it may be that by mimicking the warning colours of venomous snakes some non-venomous species can avoid predators such as birds (Brodie & Janzen, 1995), which will help to reduce extrinsic mortality, and hence species that do not possess venom may evolve longer lifespans. Finally, venom, basically a mechanism used to obtain prey, may simply not be an effective mechanism to deter potential predators. Our dataset enabled us to perform the largest (albeit still only ~12% of known reptilian diversity) comparative study of ectotherm longevity carried out to date. The large sample size, great spatial scope and the broad taxonomic sampling enabled us to observe patterns that would not have been feasible with small-scale studies. This type of study, however, can only suggest mechanisms based on correlations and not experimentally support (or refute) them. Experimental manipulations are not feasible at this scale, or with lifespans that can span decades. It is, of course, unethical to perform such manipulations, even if they had been possible. Thus, comparative methods may be the best way of conducting such studies (e.g. Wilkinson & South, 2002; Magalhães et al., 2007; Healy et al., 2014; Valcu et al., 2014). The low explanatory power we achieve, however, could reflect issues with underlying data (see discussion in Scharf et al., 2015), or with the nature and coarseness of our predictor variables. In conclusion, some of our results support the evolutionary hypotheses of senescence that were proposed to explain the variation in longevity for different animal species (Medawar, 1952; Hamilton, 1966; Kirkwood, 1977; Kirkwood and Austad, 2000). Large size, insularity, low temperatures and high seasonality all promote reptile longevity. Some factors related to both extrinsic and intrinsic mortality rates, such as use of venom for defense and activity period, however, emerged as not affecting longevity in reptiles. We think that obtaining more direct measures of extrinsic mortality generators may prove illuminating regarding the merit of the evolutionary theory of senescence in reptiles and in ectotherms in general. SUPPORTING INFORMATION Additional Supporting Information may be found in the online version of this article at the publisher’s web-site. Appendix S1. Longevity database. Table S1.2. Longevity references. Appendix S2. Crocodilia tree reconstruction. Appendix S3. Full models and sensitivity analysis results. Appendix S3.1. Body size and longevity. Appendix S3.2. Island area. Appendix S3.3. Full models. Appendix S3.4. Models with data from wild individuals only. Appendix S3.5. Well-sampled analysis – Minimum Adequate Model (MAM). Appendix S3.6. Summary of results. Appendix S3.7. Models for snakes and turtles. Appendix S3.8. Origin of data - t-test. ACKNOWLEDGEMENTS We thank Rachel Schwarz and Simon Jameson for constructive discussion and Maria Novosolov for assistance with statistical modelling. We thank Inon Scharf, Kevin Healy and three anonymous referees for insightful comments on earlier drafts of the manuscript. Data availability statement: all data used in the analyses and the associated metadata are available in Appendix S1. REFERENCES Adler GH , Levins R . 1994 . The island syndrome in rodent populations . The Quarterly Review of Biology 69 : 473 – 490 . Google Scholar Crossref Search ADS PubMed Austad SN . 1993 . Retarded senescence in an insular population of Virginia opossums (Didelphis virginiana) . Journal of Zoology 229 : 695 – 708 . Google Scholar Crossref Search ADS Austad SN , Fischer KE . 1991 . Mammalian aging, metabolism, and ecology: evidence from the bats and marsupials . Journal of Gerontology 46 : B47 – B53 . Google Scholar Crossref Search ADS PubMed Benjamin DJ , Berger JO , Johannesson M , Nosek BA , Wagenmakers EJ , Berk R , Bollen KA , Brembs B , Brown L , Camerer C , Cesarini D . 2018 . Redefine statistical significance . Nature Human Behaviour 2 : 6 – 10 . Google Scholar Crossref Search ADS Blanco MA , Sherman PW . 2005 . Maximum longevities of chemically protected and non-protected fishes, reptiles, and amphibians support evolutionary hypotheses of aging . Mechanisms of Ageing and Development 126 : 794 – 803 . Google Scholar Crossref Search ADS PubMed Brodie ED III , Janzen FJ . 1995 . Experimental studies of coral snake mimicry: generalized avoidance of ringed snake patterns by free-ranging avian predators . Functional Ecology 9 : 186 – 190 . Google Scholar Crossref Search ADS Bronikowski A , Vleck D . 2010 . Metabolism, body size and life span: a case study in evolutionarily divergent populations of the garter snake (Thamnophis elegans) . Integrative and Comparative Biology 50 : 880 – 887 . Google Scholar Crossref Search ADS PubMed Brys K , Vanfleteren JR , Braeckman BP . 2007 . Testing the rate-of-living/oxidative damage theory of aging in the nematode model Caenorhabditis elegans . Experimental Gerontology 42 : 845 – 851 . Google Scholar Crossref Search ADS PubMed Buchanan BW . 1998 . Low-illumination prey detection by squirrel treefrogs . Journal of Herpetology 32 : 270 – 274 . Google Scholar Crossref Search ADS Buffenstein R , Jarvis JU . 2002 . The naked mole rat–a new record for the oldest living rodent . Science of Aging Knowledge Environment: SAGE KE 2002 : pe7 . Google Scholar PubMed Byrnes G , Spence AJ . 2011 . Ecological and biomechanical insights into the evolution of gliding in mammals . Integrative and Comparative Biology 51 : 991 – 1001 . Google Scholar Crossref Search ADS PubMed Caswell H . 2007 . Extrinsic mortality and the evolution of senescence . Trends in Ecology & Evolution 22 : 173 – 174 . Google Scholar Crossref Search ADS PubMed Carey JR . 2003 . Longevity: the biology and demography of life span . Princeton University Press . Daly M , Behrends PR , Wilson MI , Jacobs LF . 1992 . Behavioral modulation of predation risk: moonlight avoidance and crepuscular compensation in a nocturnal desert rodent, Dipodomys merriami . Animal Behaviour 44 : 1 – 9 . Google Scholar Crossref Search ADS Downes SJ . 2002 . Size-dependent predation by snakes: selective foraging or differential prey vulnerability ? Behavioral Ecology 13 : 551 – 560 . Google Scholar Crossref Search ADS Drummond AJ , Suchard MA , Xie D , Rambaut A . 2012 . Bayesian phylogenetics with BEAUti and the BEAST 1.7 . Molecular Biology and Evolution 29 : 1969 – 1973 . Google Scholar Crossref Search ADS PubMed Feldman A . 2015 . Snake macroecology . PhD dissertation, Tel Aviv University . Feldman A , Meiri S . 2013 . Length–mass allometry in snakes . Biological Journal of the Linnean Society 108 : 161 – 172 . Google Scholar Crossref Search ADS Feldman A , Meiri S . 2014 . Australian snakes do not follow Bergmann’s rule . Evolutionary Biology 41 : 327 – 335 . Google Scholar Crossref Search ADS Feldman A , Sabath N , Pyron RA , Mayrose I , Meiri S . 2016 . Body sizes and diversification rates of lizards, snakes, amphisbaenians and the tuatara . Global Ecology and Biogeography 25 : 187 – 197 . Google Scholar Crossref Search ADS Fick SE , Hijmans RJ . 2017 . WorldClim 2: new 1‐km spatial resolution climate surfaces for global land areas . International Journal of Climatology 37 : 4302 – 4315 . Google Scholar Crossref Search ADS Finch CE . 1990 . Senescence, longevity, and the genome . Chicago : University of Chicago Press . Freckleton RP , Harvey PH , Pagel M . 2002 . Phylogenetic analysis and comparative data: a test and review of evidence . The American Naturalist 160 : 712 – 726 . Google Scholar Crossref Search ADS PubMed Fry BG , Vidal N , Norman JA , Vonk FJ , Scheib H , Ramjan SF , Kuruppu S , Fung K , Hedges SB , Richardson MK , Hodgson WC , Ignjatovic V , Summerhayes R , Kochva E . 2006 . Early evolution of the venom system in lizards and snakes . Nature 439 : 584 – 588 . Google Scholar Crossref Search ADS PubMed Grant P . 1998 . Patterns on islands and microevolution . In: Grant PR , ed. Evolution on islands . Oxford : Oxford University Press , 1 – 17 . Hamilton WD . 1966 . The moulding of senescence by natural selection . Journal of Theoretical Biology 12 : 12 – 45 . Google Scholar Crossref Search ADS PubMed Healy K , Guillerme T , Finlay S , Kane A , Kelly SBA , McClean D , Kelly DJ , Donohue I , Jackson AL , Cooper N . 2014 . Ecology and mode-of-life explain lifespan variation in birds and mammals . Proceedings of the Royal Society of London B: Biological Sciences 281 : 20140298 . Google Scholar Crossref Search ADS Holm S , Davis RB , Javoiš J , Õunap E , Kaasik A , Molleman F , Tammaru T . 2016 . A comparative perspective on longevity: the effect of body size dominates over ecology in moths . Journal of Evolutionary Biology 29 : 2422 – 2435 . Google Scholar Crossref Search ADS PubMed Holmes DJ , Austad SN . 1994 . Fly now, die later: life-history correlates of gliding and flying in mammals . Journal of Mammalogy 75 : 224 – 226 . Google Scholar Crossref Search ADS Hossie TJ , Hassall C , Knee W , Sherratt TN . 2013 . Species with a chemical defence, but not chemical offence, live longer . Journal of Evolutionary Biology 26 : 1598 – 1602 . Google Scholar Crossref Search ADS PubMed Itescu Y . 2012 . A biogeographic perspective on turtle evolution . MSc thesis, Tel Aviv University . Itescu Y , Karraker NE , Raia P , Pritchard PC , Meiri S . 2014 . Is the island rule general? Turtles disagree . Global Ecology and Biogeography 23 : 689 – 700 . Google Scholar Crossref Search ADS Jansa SA , Voss RS . 2011 . Adaptive evolution of the venom-targeted vWF protein in opossums that eat pitvipers . PLoS ONE 6 : e20997 . Google Scholar Crossref Search ADS PubMed Johnson VE . 2013 . Revised standards for statistical evidence . Proceedings of the National Academy of Sciences USA 110 : 19313 – 19317 . Google Scholar Crossref Search ADS Johnston IA , Dunn J . 1987 . Temperature acclimation and metabolism in ectotherms with particular reference to teleost fish . Symposia of the Society for Experimental Biology 41 : 67 – 93 . Google Scholar PubMed Jordana X , Marín-Moratalla N , DeMiguel D , Kaiser TM , Köhler M . 2012 . Evidence of correlated evolution of hypsodonty and exceptional longevity in endemic insular mammals . Proceedings of the Royal Society of London B: Biological Sciences 279 : 3339 – 3346 . Google Scholar Crossref Search ADS Kirkwood TB . 1977 . Evolution of ageing . Nature 270 : 301 – 304 . Google Scholar Crossref Search ADS PubMed Kirkwood TB , Austad SN . 2000 . Why do we age ? Nature 408 : 233 – 238 . Google Scholar Crossref Search ADS PubMed Kirkwood TB , Rose MR . 1991 . Evolution of senescence: late survival sacrificed for reproduction . Proceedings of the Royal Society of London B: Biological Sciences 332 : 15 – 24 . Google Scholar Crossref Search ADS Koopman JJ , Wensink MJ , Rozing MP , van Bodegom D , Westendorp RG . 2015 . Intrinsic and extrinsic mortality reunited . Experimental Gerontology 67 : 48 – 53 . Google Scholar Crossref Search ADS PubMed Lindstedt SL , Calder WA III . 1981 . Body size, physiological time, and longevity of homeothermic animals . The Quarterly Review of Biology 56 : 1 – 16 . Google Scholar Crossref Search ADS Magalhães JPD , Costa J , Church GM . 2007 . An analysis of the relationship between metabolism, developmental schedules, and longevity using phylogenetic independent contrasts . The Journals of Gerontology Series A: Biological Sciences and Medical Sciences 62 : 149 – 160 . Google Scholar Crossref Search ADS Martín J , Marcos I , López P . 2005 . When to come out from your own shell: risk-sensitive hiding decisions in terrapins . Behavioral Ecology and Sociobiology 57 : 405 – 411 . Google Scholar Crossref Search ADS Medawar PB . 1952 . An unsolved problem of biology . London : H. K. Lewis . Meiri S . 2007 . Size evolution in island lizards . Global Ecology and Biogeography 16 : 702 – 708 . Google Scholar Crossref Search ADS Meiri S . 2008 . Evolution and ecology of lizard body sizes . Global Ecology and Biogeography 17 : 724 – 734 . Google Scholar Crossref Search ADS Meiri S . 2010 . Length–weight allometries in lizards . Journal of Zoology 281 : 218 – 226 . Meiri S . 2018 . Traitsof lizards of the world: variation around a successful evolutionary design . Global Ecology and Biogeography , doi: https://doi.org/10.1111/geb.12773. Meiri S , Bauer AM , Chirio L , Colli GR , Das I , Doan TM , Feldman A , Herrera FC , Novosolov M , Pafilis P , Pincheira‐Donoso D . 2013 . Are lizards feeling the heat? A tale of ecology and evolution under two temperatures . Global Ecology and Biogeography 22 : 834 – 845 . Google Scholar Crossref Search ADS McCue MD . 2006 . Cost of producing venom in three North American pitviper species . Copeia 2006 : 818 – 825 . Google Scholar Crossref Search ADS Mori A , Burghardt GM , Savitzky AH , Roberts KA , Hutchinson DA , Goris RC . 2012 . Nuchal glands: a novel defensive system in snakes . Chemoecology 22 : 187 – 198 . Google Scholar Crossref Search ADS Munch SB , Salinas S . 2009 . Latitudinal variation in lifespan within species is explained by the metabolic theory of ecology . Proceedings of the National Academy of Sciences, USA 106 : 13860 – 13864 . Google Scholar Crossref Search ADS Novosolov M , Raia P , Meiri S . 2013 . The island syndrome in lizards . Global Ecology and Biogeography 22 : 184 – 191 . Google Scholar Crossref Search ADS Oaks JR . 2011 . A time-calibrated species tree of Crocodylia reveals a recent radiation of the true crocodiles . Evolution 65 : 3285 – 3297 . Google Scholar Crossref Search ADS PubMed O’Brien RM . 2007 . A caution regarding rules of thumb for variance inflation factors . Quality & Quantity 41 : 673 – 690 . Google Scholar Crossref Search ADS Orme CDL , Freckleton RP , Thomas GH , Petzoldt T , Fritz SA , Isaac NJB . 2013 . Caper: comparative analyses of phylogenetics and evolution in R. R package version 0.5 . Available at: http://CRAN.R-project.org/package=caper. Pereira AG , Sterli J , Moreira FRR , Schrago CG . 2017 . Multilocus phylogeny and statistical biogeography clarify the evolutionary history of major lineages of turtles . Molecular Phylogenetics and Evolution 113 : 59 – 66 . Google Scholar Crossref Search ADS PubMed Pianka ER . 2011 . Notes on the natural history of the tiny skink Menetia greyi . Western Australian Naturalist 28 : 12 – 17 . Quinlan RJ . 2010 . Extrinsic mortality effects on reproductive strategies in a Caribbean community . Human Nature 21 : 124 – 139 . Google Scholar Crossref Search ADS Robinson JE , St John FA , Griffiths RA , Roberts DL . 2015 . Captive reptile mortality rates in the home and implications for the wildlife trade . PloS ONE 10 : e0141460 . Google Scholar Crossref Search ADS PubMed Roll U , Feldman A , Novosolov M , Allison A , Bauer AM , Bernard R , Böhm M , Castro-Herrera F , Chirio L , Collen B , Colli GR , Dabool L , Das I , Doan TM , Grismer LL , Hoogmoed M , Itescu Y , Kraus F , LeBreton M , Lewin A , Martins M , Maza E , Meirte D , Nagy ZT , de C Nogueira C , Pauwels OSG , Pincheira-Donoso D , Powney GD , Sindaco R , Tallowin OJS , Torres-Carvajal O , Trape JF , Vidan E , Uetz P , Wagner P , Wang Y , Orme CDL , Grenyer R , Meiri S . 2017 . The global distribution of tetrapods reveals a need for targeted reptile conservation . Nature Ecology & Evolution 1 : 1677 – 1682 . Google Scholar Crossref Search ADS PubMed Roquet C , Lavergne S , Thuiller W . 2014 . One tree to link them all: a phylogenetic dataset for the European Tetrapoda . PLoS Currents 6 : pii . Scharf I , Feldman A , Novosolov M , Pincheira‐Donoso D , Das I , Böhm M , Uetz P , Torres‐Carvajal O , Bauer A , Roll U , Meiri S . 2015 . Late bloomers and baby boomers: ecological drivers of longevity in squamates and the tuatara . Global Ecology and Biogeography 24 : 396 – 405 . Google Scholar Crossref Search ADS Shattuck MR , Williams SA . 2010 . Arboreality has allowed for the evolution of increased longevity in mammals . Proceedings of the National Academy of Sciences, USA 107 : 4635 – 4639 . Google Scholar Crossref Search ADS Slavenko A , Tallowin OJ , Itescu Y , Raia P , Meiri S . 2016 . Late Quaternary reptile extinctions: size matters, insularity dominates . Global Ecology and Biogeography 25 : 1308 – 1320 . Google Scholar Crossref Search ADS Smith MT , Ortega J , Beaupre SJ . 2014 . Metabolic cost of venom replenishment by Prairie Rattlesnakes (Crotalus viridis viridis) . Toxicon 86 : 1 – 7 . Google Scholar Crossref Search ADS PubMed Sohal RS . 1986 . The rate of living theory: a contemporary interpretation . In: Collatz K , Sohal RS , eds. Insect Aging . Berlin, Heidelberg : Springer . Sohal RS , Allen RG . 1986 . Relationship between oxygen metabolism, aging and development . Advances in Free Radical Biology & Medicine 2 : 117 – 160 . Google Scholar Crossref Search ADS Sparkman AM , Arnold SJ , Bronikowski AM . 2007 . An empirical test of evolutionary theories for reproductive senescence and reproductive effort in the garter snake Thamnophis elegans . Proceedings of the Royal Society of London B: Biological Sciences 274 : 943 – 950 . Google Scholar Crossref Search ADS Speakman JR . 2005 . Body size, energy metabolism and lifespan . The Journal of Experimental Biology 208 : 1717 – 1730 . Google Scholar Crossref Search ADS PubMed Stark G , Meiri S . 2018 . Coldand dark captivity: drivers of amphibian longevity . Global Ecology and Biogeography , doi: https://doi.org/10.1111/geb.12804. Tonini JFR , Beard KH , Ferreira RB , Jetz W , Pyron RA . 2016 . Fully-sampled phylogenies of squamates reveal evolutionary patterns in threat status . Biological Conservation 204 : 23 – 31 . Google Scholar Crossref Search ADS Turbill C , Bieber C , Ruf T . 2011 . Hibernation is associated with increased survival and the evolution of slow life histories among mammals . Proceedings of the Royal Society of London B: Biological Sciences 278 : 3355 – 3363 . Google Scholar Crossref Search ADS Uetz P , Freed P , Hošek J , eds. The Reptile Database . Available at: http://www.reptile-database.org. Valcu M , Dale J , Griesser M , Nakagawa S , Kempenaers B . 2014 . Global gradients of avian longevity support the classic evolutionary theory of ageing . Ecography 37 : 930 – 938 . Google Scholar Crossref Search ADS Vidan E , Roll U , Bauer A , Grismer L , Guo P , Maza E , Novosolov M , Sindaco R , Wagner P , Belmaker J , Meiri S . 2017 . The Eurasian hot nightlife: environmental forces associated with nocturnality in lizards . Global Ecology and Biogeography 26 : 1316 – 1325 . Google Scholar Crossref Search ADS Wachter KW , Finch CE . 1997 . Between Zeus and the salmon: the biodemography of longevity . Washington : The National Academies Press . Williams GC . 1957 . Pleiotropy, natural selection, and the evolution of senescence . Evolution 11 : 398 – 411 . Google Scholar Crossref Search ADS Williams SA , Shattuck MR . 2015 . Ecology, longevity and naked mole-rats: confounding effects of sociality ? Proceedings of the Royal Society of London B: Biological Sciences 282 : 20141664 . Google Scholar Crossref Search ADS Wilkinson GS , South JM . 2002 . Life history, ecology and longevity in bats . Aging Cell 1 : 124 – 131 . Google Scholar Crossref Search ADS PubMed Williams PD , Day T . 2003 . Antagonistic pleiotropy, mortality source interactions, and the evolutionary theory of senescence . Evolution 57 : 1478 – 1488 . Google Scholar Crossref Search ADS PubMed Williams PD , Day T , Fletcher Q , Rowe L . 2006 . The shaping of senescence in the wild . Trends in Ecology & Evolution 21 : 458 – 463 . Google Scholar Crossref Search ADS PubMed Zhang L , Lu XIN . 2012 . Amphibians live longer at higher altitudes but not at higher latitudes . Biological Journal of the Linnean Society 106 : 623 – 632 . Google Scholar Crossref Search ADS © 2018 The Linnean Society of London, Biological Journal of the Linnean Society This article is published and distributed under the terms of the Oxford University Press, Standard Journals Publication Model (https://academic.oup.com/journals/pages/open_access/funder_policies/chorus/standard_publication_model)
TI - Cold and isolated ectotherms: drivers of reptilian longevity
JO - Biological Journal of the Linnean Society
DO - 10.1093/biolinnean/bly153
DA - 2018-10-26
UR - https://www.deepdyve.com/lp/oxford-university-press/cold-and-isolated-ectotherms-drivers-of-reptilian-longevity-6faDfp4FbJ
SP - 1
VL - Advance Article
IS - 
DP - DeepDyve
ER -